Self-reports are a primary source of behavioral data. Studies by this group and others have highlighted the variable validity and reliability of self-report measures of health behaviors such as adherence with therapy. Research on self-reports of sensitive information, such as sexual behavior and drug use, suggests that mode of data collection may enhance validity of self-reports. However, no studies have determined how mode of data collection influences self-reports when an objective measure of the behavior is available. The goal of this study is to examine how assessment mode influences the validity of self-reported adherence with asthma therapy. Self-reported adherence with asthma therapy will be collected using one of three randomized assessment modes: interview, computer or telephone. Adult patients will be enrolled in a 6-month asthma education study where their inhaled medications are electronically monitored to provide an objective adherence measure. A total of 216 adults will be randomized to these three treatments. The investigators hypothesize that self-reports will have the highest validity under conditions of lowest personalization. They will also examine how interview mode influences self-reports of other behavioral and clinical outcome measures including asthma symptoms, asthma management and quality of life. Results from this study will demonstrate how data collection methods influence self-report of adherence with therapy and serve to identify optimal self-report strategies in behavioral and clinical research settings.